Control of Random Variation by Blocking
In a field experiment it would be possible, as a result of the randomization, for all plots with a particular treatment to be grouped together in one corner of the experiment. While in the long run the completely randomized design gives a fair comparison of treatments, a particular experiment might prove unfortunate. The logical way to overcome this possibility of an unfortunate design is to restrict the randomization in some way. The randomized block design attempts to control one form of local variation. It is worth noting that any design using blocking could have occurred by chance if blocking were not used, but the reverse is not true. The benefit from blocking comes partly from ensuring that each treatment is assessed for a similar mixture of "good" and "bad" units. As one might expect, the method of analyzing a Latin square design is similar to that for a randomized block design with one additional source of variation to be included.